| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Stroke. 2001;32:1085.)
© 2001 American Heart Association, Inc.
Original Contributions |
Diagnosis and Initial Management of Stroke and Transient Ischemic Attack Across UK Health Regions From 1992 to 1996
Experience of a National Primary Care Database
From the Department of Vascular Surgery (R.G.J.G., R.M.G., A.H.D.), Imperial College School of Medicine, Charing Cross Campus, London, UK; the Department of Public Health Sciences (R.N.), Guy’s, King’s and St Thomas’ School of Medicine, Capital House, Guy’s Hospital, London, UK; and Postgraduate Medical School (R.L.), University of Surrey, Stirling House Campus, Surrey Research Park, Guildford, Surrey, UK.
Correspondence to R.G.J. Gibbs, Research Fellow, Department of Vascular Surgery, Imperial College School of Medicine, Charing Cross Campus, Fulham Palace Road, London W6 8RF, UK. E-mail richard.gibbs{at}virgin.net
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Background and Purpose—The aim of this study was to establish the difference in burden of cerebrovascular disease across the different health regions of the United Kingdom and to determine whether the initial management of new cases of stroke and transient ischemic attack (TIA) was uniform across the United Kingdom.
Methods—The General Practice Research Database (GPRD) is a national database used for epidemiological studies. This was a cohort study identifying incident cases of stroke and TIA over a 5-year study period between 1992 and 1996. The population studied was patients registered with general practitioners contributing to the GPRD across the different health regions of the United Kingdom. Outcome measures were new diagnoses of stroke and TIA, new prescriptions for antiplatelet and anticoagulant agents, and referrals made for specialist opinion.
Results—The age-adjusted annual incidence rate across all regions was 151 per 100 000 for stroke and 190 per 100 000 for TIA. There was almost a 2-fold difference in the incidence of cerebrovascular disease between the regions. The management of stroke and TIA in terms of antiplatelet prescription and of referral onward for further opinion to hospital specialists varied significantly between regions.
Conclusions—Reported stroke and TIA incidence on the GPRD was comparable to that of other European studies. There were striking regional differences in the incidence of disease. The primary care management, both in prescription and referral rates, varied significantly between the different regions. There was a marked underuse of antiplatelet and anticoagulant agents, and referral rates for specialists’ opinions were low.
Key Words: cerebral ischemia, transient • epidemiology • Great Britain • stroke, ischemic • stroke management
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Stroke is a common cause of morbidity and mortality in developed countries,1 with 100 000 new cases of stroke per year in England and Wales.2 Emphasis has been placed on strategies for prevention by medical and surgical interventions.3 Both primary and secondary preventative strategies consume healthcare resources, and it has been suggested that there is unmet need in the provision of services, particularly for the surgical procedure of carotid endarterectomy.4 For an equitable allocation of resources to occur, the geographical variation in burden of disease must be taken into account. The initial management of transient ischemic attack (TIA) and ischemic stroke is essential in the secondary prevention of further stroke, and trends in prescribing and referral are indicators of how cerebrovascular disease is dealt with in the primary care setting.
The General Practitioner Research Database (GPRD) is a UK database that has been used in primary care epidemiological studies5 and is recognized as a potentially useful source of national morbidity data.6 All patients in the care of a participating general practitioner (GP) are automatically registered onto the GPRD. Participating GPs enter clinical information, including significant diagnoses and subsequent interventions, such as prescriptions and referrals. Patients diagnosed and managed at home are recorded; a validation carried out in 1991 suggested that 87% of patients seen in the hospital had their diagnoses recorded on the database.7 Thus, the completeness of recording is high. We used this database to investigate the frequency of diagnosis of cerebrovascular disease as perceived by GPs and to determine whether the initial management of stroke and TIA was uniform across the different Regional Health Authorities (RHAs) of the United Kingdom. This work obtained the necessary approval of the Scientific and Ethics Advisory Group of the Office for National Statistics.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Data for patients with new cases of stroke and TIA presenting to GPs were retrieved from the GPRD by using appropriate diagnostic codes (see Table 1
). Diagnostic codes are entered
onto the database by using the Oxford medical information system
(OXMIS) codes. This coding system is derived from the eighth revision
of the International Classification of Diseases. The OXMIS codes are
designed to be all-encompassing, giving a wide choice of codes for the
same condition to ensure that a code matching the GP’s diagnosis is
available. The database contains complete information on age, sex, and
prescribing data. There are incomplete data on weight, blood pressure,
and smoking habits and on diagnostic tests, whereas a
record of hospitalization is an administrative procedure that is
well recorded.
|
The study period ran for 5 consecutive years between 1992 and 1996. Patients with a diagnosis of stroke or TIA from preceding years were excluded to ensure that only new subject data were retrieved for analysis. Prescriptions for antiplatelet or anticoagulant agents were analyzed, and all strengths from the minimum to the maximum were included. The antiplatelet agents that were searched for were aspirin (all strengths from 30 to 648 mg) and dipyridamole (25 to 100 mg), and the anticoagulant agents that were searched for were warfarin (1 to 5 mg), nicoumalone (1 to 4 mg), and phenindione (10 to 50 mg). Any referrals made for specialists’ opinions within 7 days of the new diagnostic code entry were analyzed.
Statistical analyses were undertaken by use of Stata
(StataCorp, 1999) and SAS version 6 (SAS Institute, 1989). The crude
incidence data for each region was standardized by 5-year age group for
the UK population in 1996. Correlations between regional rates of
incidence, referral, and antiplatelet prescription was measured by
Kendall 
-a, with 95% confidence
limits8 and
continuity-corrected probability values.
Sixty case notes with a diagnostic code for TIA
(30 cases) and stroke (30 cases) for patients who had been referred for
specialists’ opinions were obtained randomly and anonymously from
participating practices. Depersonalized copies of all correspondence
from the specialist to whom the patient had been referred were
examined. The diagnosis made by the specialist was compared with the
diagnostic code entered onto the GPRD by the GP to
determine how often they were in agreement, as a test of the validity
of the diagnostic data entered onto the GPRD (see
Table 2).
|
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The GPRD had a mean population of 2 442 911 between 1992 and 1996, and we retrieved information on 17 404 patients with a new diagnosis of stroke and 22 195 patients with a new diagnosis of TIA.
Crude mean annual stroke incidence per 100 000 was 143,
age-adjusted to 151 for the UK population of 1996. Crude TIA incidence
was 182, age-adjusted to 190. The incidence of both TIA and stroke
varied between the different health regions (see
Table 3). The 
2 test was
used to test for differences in incidence of stroke and TIA between the
RHAs. Differences between RHAs in stroke incidence were statistically
significant when pooled between years and also for each year
(P<0.00005 in all cases).
Statistically significant differences in the incidence of TIA for the
pooled years and for each individual year were also observed between
the RHAs (P<0.00005 in all
cases). Yorkshire had the highest incidence, at 194 (stroke) and 249
(TIA) per 100 000 of the population, compared with 115 and 122 per
100 000 for Oxford, the region with the lowest incidence.
|
Prescription rates for antiplatelet agents varied
significantly between the RHAs for both TIA and stroke patients.
Antiplatelets were prescribed for 15% to 25% of the stroke
patients and for 30% to 45% of the TIA patients over the 5 years,
depending on region (see
Figure 1). Over time, the prescription of
antiplatelets fluctuated 
37.5% for TIA and 17% for stroke.
Prescription rates for anticoagulants were 1.26% (95% CI 0.97% to
1.47%) for stroke and 1.04% (95% CI 0.72% to 1.56%) for TIA for
the time period.
|
The variation in referral rates between RHAs was
statistically significant; they were pooled over all years and within
each year from 1992 to 1996 and included both stroke and TIA. Referral
rates varied from 14% (Yorkshire) to 26% (Oxford) for TIA, with a
mean of 18.8% over the 5-year period for all regions. For stroke,
referral rates varied more strikingly between RHAs, from 30% in
Yorkshire to 56% in Scotland, with a mean of 39% (see
Figure 2).
|
We hypothesized that the referral rate would be related to
the local incidence of disease; if the regional incidence was high,
there would be greater expertise in managing stroke and TIA among GPs,
resulting from increased experience and, possibly, more accessible
referral pathways to secondary stroke care. The correlation between
incidence and referral rates was examined. These were nonsignificant
for stroke (Kendall -a=-0.25, 95% CI -0.64 to 0.25;
P=0.1740) and for TIA (Kendall
-a=-0.35, 95% CI -0.69 to 0.11;
P=0.0529). The correlation
between referral rate and antiplatelet prescription rate was also
examined to determine whether GPs were prescribing initial drug therapy
as the sole modality of treatment and not referring the patient onward.
These are nonsignificant for stroke (Kendall -a=-0.28, 95% CI
-0.66 to 0.21; P=0.1275) and
for TIA (Kendall -a=-0.10, 95% CI -0.51 to 0.34;
P=0.5923).
To validate the GP diagnostic coding, 60 patient records were obtained. Eight records had insufficient information and were excluded. For the 27 patients referred with a diagnostic code for TIA, the GP and specialist concurred exactly in 13 cases (48%), and a diagnosis of stroke rather than TIA was made by the specialist in a further 5 cases (18%). This gave an overall agreement in opinion of cerebrovascular diagnosis of 66%. For the 25 referrals with a diagnostic code for stroke, there was a concurrence of opinion in exactly in 16 cases (64%), and a diagnosis of TIA rather than stroke was made in a further 4 cases (16%), giving an overall agreement in opinion of cerebrovascular diagnosis of 80%.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The mean annual incidence of stroke was 151 per 100 000 within this national database over the 5-year study period, and there was a significant variance in reporting of new cases of cerebrovascular disease between the different health regions.
The incidence data from the GPRD for stroke is comparable to figures reported from European stroke registers. Incidence rates of stroke reported range from 111 per 100 000 from Warsaw,9 145 per 100 000 from Tilburg, the Netherlands,10 and 155 per 100 000 from Umbria11 up to 170 per 100 000 from Belluno, Italy,12 and 221 per 100 000 from Innherred, Norway.13 The Oxfordshire Community Stroke Project reported a crude incidence rate of 160 per 100 000 adjusted to 200 per 100 000.14 Reporting of new cases of stroke to the GPRD in the Oxford region gave a crude incidence rate of 95 per 100 000, adjusted to 131 per 100 000 for the 1996 population. Direct comparison is not possible; the populations sampled were different, study methodology was different, and the studies are a decade apart. Nonetheless, comparing the incidence rates from different studies of very similar populations raises the issue of the validity of the data reported to the GPRD.
There are a number of methodological difficulties in using a very large database covering a wide geographical area. The first problem is the definition of stroke for selecting subjects for inclusion in the study. Most studies use the World Health Organization definition of stroke,15 but that was not possible in the present study. Instead, we relied on the OXMIS coding system, which is designed to ensure that participating GPs are able to match their own diagnosis with an appropriate code. The aim of this system is to ensure that cases are not missed, although the specificity of a single definition is lost. Moreover, we included only those codes appropriate for ischemic stroke. Most studies include subarachnoid and intracerebral hemorrhages in their figures, but we felt that the inclusion of codes for headache and meningism would lead to an unacceptable overestimation of stroke incidence.
True incident cases must be first-ever strokes. Attention was paid to the issue of case ascertainment in the methodology. To minimize the possibility of counting recurrent strokes, once a diagnostic code for stroke or TIA was entered, that patient was subsequently excluded for the rest of the study period. In addition, any patient with a diagnostic code for cerebrovascular disease from the inception of the GPRD in 1987 and before the study start date was also excluded. It is possible that patients could potentially have had strokes before the start of the database, and if that diagnosis had not been entered, then a recurrent stroke would be counted as a first-ever stroke. However, contributing GPs would be expected to enter major clinical events such as stroke in a retrospective as well as contemporaneous fashion, but these data are not completely reliable. Thus, although we were able to reliably count first strokes since the start of the study, it is possible that these were not first-in-a-lifetime strokes, the criterion for incident cases.
Community-based studies are more accurate than are hospital-based studies because the rates of hospital admission for stroke vary on the basis of both location and time.16 This is even more relevant in the case of TIA and minor stroke, which can be managed in the community. The GPRD captured all patients presenting either primarily to the GP or secondarily on discharge from hospital back to the care of the GP. If patients bypassed the GP by presenting to the hospital directly, the diagnostic code would be entered subsequently, on receiving the hospital discharge summary. If patients were admitted to the hospital and subsequently died, they were removed from the doctor’s list, and the cause of death was usually recorded on the database. Hence, the GP data are the most complete data available from routine sources of new events because these data include patients with episodes that are diagnosed and managed in the community as well as patients that are seen in the hospital.
Both the size and the nature of the population studied influence estimates of disease incidence.17 The population of the GPRD increased year on year, from 1 816 068 in 1992 to 2 685 432 in 1996. In general terms, the larger the population studied, the greater is the number of incident cases that will occur, and the narrower are the CIs. The nature of the population studied is important; studies of older populations will report higher incidences than will studies of the general population. The GPRD population consists of people registered with participating practices and, as such, should be representative of the population at large.
The mean annual incidence of TIA reported over the 5-year study period was 190 per 100 000. There are fewer reports of TIA incidence compared with stroke incidence in the literature. Studies attempting to measure incidence rates for TIA in different populations report widely differing rates, and this again depends on the methodology used. One of the difficulties in accurately estimating the incidence of TIA is the transitory nature of the symptoms, which makes patients less likely to seek medical advice.18 Therefore, studies relying on case ascertainment through routine medical sources such as general practice and hospital records may be underestimating TIA rates. In an attempt to address this point, several studies have used survey methodologies with direct questions on symptomatology. Prospective registry-type studies with stringent inclusion criteria tend to report lower incidences, ranging from 50 per 100 00019 to 80 per 100 000,20 21 whereas studies relying on survey methodologies report higher incidence rates, ranging from 280 per 100 00022 to 310 per 100 000.23 It may be that the true incidence of TIA is likely to lie somewhere between the high incidence reported by subjects in the community when directly questioned and the low incidence reported by prospective observational studies.
The higher reported incidence of TIA compared with stroke
has several possible explanations. The clinical diagnosis of TIA is
recognized to be difficult, with poor interobserver
agreement,24 and the
distinction between stroke and TIA is dependent on time. If the
diagnostic code of TIA is entered onto the GPRD, it is
possible that the symptoms could persist for >24 hours without the GP
being aware of the fact, which would lead to an underestimation of
stroke and an overestimation of TIA. Although there is the facility for
entry codes to be altered subsequently, this would not necessarily
happen. In the present study, reported TIA rates were uniformly
higher than stroke rates, possibly reflecting this methodological
problem. The case notes that were examined to verify the accuracy of
diagnosis made by study GPs confirm this problem.
Diagnostic agreement between the study GPs and the
specialists to whom the stroke patients were referred reached 64%. A
further 16% of patients initially diagnosed with stroke by the GP were
subsequently given the diagnosis of TIA by the specialist, presumably
with the additional benefit of time. The same pattern was repeated for
patients given a diagnostic code for TIA by the GP, with
18% of these cases subsequently diagnosed as stroke by the specialist.
The validation exercise also illustrates the difficulty in obtaining
accurate diagnoses of TIAs by GPs. The agreement in diagnosis between
study GPs and specialists reached only 48%, and in a further 33% of
cases, the final diagnoses reached by the specialists were not even
cerebrovascular in etiology (see
Table 2
).
Antiplatelet agents are recognized prophylaxes for patients with TIA and stroke in the prevention of further adverse vascular events,25 and there is an overall risk reduction of 4.3% in patients who are prescribed aspirin for previous stroke or TIA.26 Despite this, there is some evidence that there is suboptimal secondary prevention in patients suffering from stroke27 as well as other vascular diseases28 in the United Kingdom. Data for GP prescription of antiplatelets in the present study suggest that these agents are markedly underused in the secondary prevention of stroke in the United Kingdom, with prescribing rates of 37% for TIA and 17% for stroke. The variation in prescribing rates across the regions shows that even this suboptimal management is not uniform between the regions. The low overall prescription rates could partially be explained by the study methodology. Only new prescriptions issued within 7 days of first consultation and entry of diagnostic code were counted. This would have excluded patients already on antiplatelet agents for coexisting conditions and also patients returning from the hospital who had already been issued a prescription. Nonprescription aspirin obtained from the chemist would not be recorded on the database, and this would particularly affect the group aged <60 years, for whom the GPs might try to avoid prescription charges. Nonetheless, this does not explain regional differences in prescribing over the study period; the possible underestimate of prescribing would apply uniformly to all regions.
Adjusted-dose warfarin is also recognized as being useful in
the prevention of stroke in patients with nonvalvular atrial
fibrillation. The prescribing rates for anticoagulants were low: 1.26%
for stroke and 1.04% for TIA. Although the diagnostic
codes are not specific for cardioembolic events secondary to atrial
fibrillation, it could be reasonably expected that 5% of the
subjects had suffered this type of event, on the basis of the
prevalence of atrial fibrillation in the UK population. Previous
studies have found that anticoagulants are underused in the management
of atrial fibrillation,29 and
these data support this observation.
Referral rates for further opinion were higher for stroke patients (39%) than for TIA patients (18.8%). A proportion of these patients would have returned to the care of the GP from the hospital setting and would, thus, have not required referral, leading to an underestimate of patients being dually managed. However, there was again a marked difference across the regions for both TIA and stroke. One hypothesis to explain this observation is that the prescription of prophylaxis in the form of antiplatelets would constitute the totality of treatment, and further investigation or opinion would be unnecessary. However, there was no significant correlation found between regions with high prescribing rates and low referral rates.
In terms of the relationship between disease incidence and the rate of referral, generally, the regions with higher incidence rates had the lower referral rates (ie, the Yorkshire region). This tendency suggests either that health care rationing for stroke patients is occurring or that GPs who are exposed to greater numbers of patients presenting with stroke and TIA feel concomitantly more confident in dealing with them alone. The referral rates suggest that too few patients with a new diagnosis of cerebrovascular disease are being appropriately referred onward for further evaluation and management. There was a significant variation in referral across the different regions, and access to a specialist’s opinion would appear to be partially dependent on location. Taken in conjunction with the low prescribing rates, these data suggest that greater educational input is required to improve the secondary prevention of stroke and to refer greater numbers of these patients for specialists’ opinions in the United Kingdom.
|
Acknowledgments |
|---|
This work was partially funded by Sanofi Winthrop and Bristol-Myers Squibb.
Received August 7, 2000; revision received January 29, 2001; accepted January 30, 2001.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
1. Bonita R. Epidemiology of stroke. Lancet. 1992;339:342–344.[Medline] [Order article via Infotrieve]
2. Anonymous. Incidence of stroke in Oxfordshire: first year’s experience of a community stroke register. BMJ. 1983;287:713–717.
3.
Barnett HJ, Eliasziw
M, Meldrum HE. Evidence based cardiology: prevention of
ischaemic stroke. BMJ. 1999;318:1539–1543.
4.
Ferris G, Roderick
P, Smithies A, George S, Gabbay J, Couper N, Chant A. An
epidemiological needs assessment of carotid
endarterectomy in an English health region: is the
need being met? BMJ. 1998;317:447–451.
5.
Hansell A, Hollowell
J, Nichols T, McNiece R, Strachan D. Use of the General Practice
Research Database (GPRD) for respiratory
epidemiology: a comparison with the 4th
Morbidity Survey in General Practice (MSGP4).
Thorax. 1999;54:413–419.
6. Hollowell J. The General Practice Research Database: quality of morbidity data. Popul Trends. 1997;87:36–40.
7. Jick H, Jick SS, Derby LE. Validation of information recorded on general practitioner based computerised data resource in the United Kingdom. BMJ. 1991;302:766–768.
8. Newson R. Confidence intervals for "non-parametric" statistics and their differences. Stata Tech Bull. 2000;55:1547–1555.
9. Czlonkowska A, Ryglewicz D, Weissbein T, Baranska-Gieruszczak M, Hier DB. A prospective community-based study of stroke in Warsaw, Poland. Stroke. 1994;25:547–551.[Abstract]
10.
Herman B, Leyten
AC, van Luijk JH, Frenken CW, Op de Coul AA, Schulte BP.
Epidemiology of stroke in Tilburg, the
Netherlands: the population-based stroke incidence register, 2:
incidence, initial clinical picture and medical care, and three-week
case fatality. Stroke. 1982;13:629–634.
11.
Ricci S, Celani
MG, Guercini G, Rucireta P, Vitali R, La Rosa F, Duca E, Ferraguzzi R,
Paolotti M, Seppoloni D. First-year results of a community-based study
of stroke incidence in Umbria, Italy.
Stroke. 1989;20:853–857.
12.
Lauria G, Gentile
M, Fassetta G, Casetta I, Agnoli F, Andreotta G, Barp C, Caneve G,
Cavallaro A, Cielo R. Incidence and prognosis of stroke in the Belluno
province, Italy: first-year results of a community-based study.
Stroke. 1995;26:1787–1793.
13.
Ellekjaer H,
Holmen J, Indredavik B, Terent A. Epidemiology
of stroke in Innherred, Norway, 1994 to 1996: incidence and 30-day
case-fatality rate. Stroke. 1997;28:2180–2184.
14.
Bamford J,
Sandercock P, Dennis M, Warlow C, Jones L, McPherson K, Vessey M,
Fowler G, Molyneux A, Hughes T. A prospective study of acute
cerebrovascular disease in the community: the Oxfordshire Community
Stroke Project 1981–86, 1: methodology, demography and incident
cases of first-ever stroke. J Neurol
Neurosurg Psychiatry. 1988;51:1373–1380.
15. Hatano S. Experience from a multicentre stroke register: a preliminary report. Bull World Health Organ. 1976;54:541–553.[Medline] [Order article via Infotrieve]
16. Bamford J, Sandercock P, Warlow C, Gray M. Why are patients with acute stroke admitted to hospital? BMJ. 1986;292:1369–1372.
17.
Sudlow CL, Warlow
CP. Comparing stroke incidence worldwide: what makes studies
comparable? Stroke. 1996;27:550–558.
18.
Dennis MS, Bamford
JM, Sandercock PA, Warlow CP. Incidence of transient ischemic
attacks in Oxfordshire, England.
Stroke. 1989;20:333–339.
19.
Lemesle M, Milan
C, Faivre J, Moreau T, Giroud M, Dumas R. Incidence trends of
ischemic stroke and transient ischemic attacks in a
well-defined French population from 1985 through 1994.
Stroke. 1999;30:371–377.
20.
Sempere AP, Duarte
J, Cabezas C, Claveria LE. Incidence of transient ischemic
attacks and minor ischemic strokes in Segovia, Spain.
Stroke. 1996;27:667–671.
21. Lauria G, Gentile M, Fassetta G, Casetta I, Agnoli F, Andreotta G, Barp C, Caneve G, Cavallaro A, Cielo R, et al. Incidence of transient ischemic attacks in the Belluno Province, Italy: first-year results of a community-based study. Acta Neurol Scand. 1996;93:291–296.[Medline] [Order article via Infotrieve]
22. Matias-Guiu J, Oltra A, Falip R, Martin R, Galiano L. Occurrence of transient ischemic attacks in Alcoi: descriptive epidemiology. Neuroepidemiology. 1994;13:34–39.[Medline] [Order article via Infotrieve]
23. Fratiglioni L, Arfaioli C, Nencini P, Ginanneschi A, Iaquinta L, Marchi M, Inzitari D. Transient ischemic attacks in the community: occurrence and clinical characteristics: a population survey in the area of Florence, Italy. Neuroepidemiology. 1989;8:87–96.[Medline] [Order article via Infotrieve]
24.
Kraaijeveld CL,
van Gijn J, Schouten HJ, Staal A. Interobserver agreement for the
diagnosis of transient ischemic attacks.
Stroke. 1984;15:723–725.
25.
Anonymous.
Collaborative overview of randomised trials of antiplatelet
therapy, I: prevention of death, myocardial infarction, and stroke by
prolonged antiplatelet therapy in various categories of patients:
Antiplatelet Trialists’ Collaboration [published erratum appears
in BMJ. 1994;308:1540].
BMJ.
1994;308:81–106.
26. Eccles M, Freemantle N, Mason J. North of England evidence based guideline development project: guideline on the use of aspirin as secondary prophylaxis for vascular disease in primary care: North of England Aspirin Guideline Development Group [published erratum appears in BMJ. 1998;316:1733]. BMJ. 1998;316:1303–1309.
27. Mead GE, Murray H, McCollum CN, O’Neill PA. How do general practitioners manage patients at risk from stroke? Br J Clin Pract. 1996;50:426–430.[Medline] [Order article via Infotrieve]
28.
Khunti K, Sorrie
R, Jennings S, Farooqi A. Improving aspirin prophylaxis after
myocardial infarction in primary care: collaboration in multipractice
audit between primary care audit group and health authority.
BMJ. 1999;319:297.
29. Sudlow M, Thomson R, Thwaites B, Rodgers H, Kenny RA. Prevalence of atrial fibrillation and eligibility for anticoagulants in the community. Lancet. 1998;352:1167–1171.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  O. C. Sheehan, A. Merwick, L. A. Kelly, N. Hannon, M. Marnane, L. Kyne, P. M.E. McCormack, J. Duggan, A. Moore, J. Moroney, et al. Diagnostic Usefulness of the ABCD2 Score to Distinguish Transient Ischemic Attack and Minor Ischemic Stroke From Noncerebrovascular Events: The North Dublin TIA Study Stroke, November 1, 2009; 40(11): 3449 - 3454. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J Joubert, C Reid, D Barton, T Cumming, A McLean, L Joubert, J Barlow, D Ames, and S Davis Integrated care improves risk-factor modification after stroke: initial results of the Integrated Care for the Reduction of Secondary Stroke model J. Neurol. Neurosurg. Psychiatry, March 1, 2009; 80(3): 279 - 284. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Wettermark, A. Persson, and M. von Euler Secondary Prevention in a Large Stroke Population: A Study of Patients' Purchase of Recommended Drugs Stroke, October 1, 2008; 39(10): 2880 - 2885. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. J Douglas and L. Smeeth Exposure to antipsychotics and risk of stroke: self controlled case series study BMJ, August 28, 2008; 337(aug28_2): a1227 - a1227. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R Luengo-Fernandez, J Leal, A Gray, S Petersen, and M Rayner Cost of cardiovascular diseases in the United Kingdom Heart, October 1, 2006; 92(10): 1384 - 1389. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Niewada, M. Skowronska, D. Ryglewicz, B. Kaminski, A. Czlonkowska, and on behalf of the Polish National Stroke Prevention Acute Ischemic Stroke Care and Outcome in Centers Participating in the Polish National Stroke Prevention and Treatment Registry Stroke, July 1, 2006; 37(7): 1837 - 1843. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Smeeth, S. L. Thomas, A. J. Hall, R. Hubbard, P. Farrington, and P. Vallance Risk of Myocardial Infarction and Stroke after Acute Infection or Vaccination N. Engl. J. Med., December 16, 2004; 351(25): 2611 - 2618. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. K. Kapral, A. Laupacis, S. J. Phillips, F. L. Silver, M. D. Hill, J. Fang, J. Richards, J. V. Tu, and for the Investigators of the Registry of the Canad Stroke Care Delivery in Institutions Participating in the Registry of the Canadian Stroke Network Stroke, July 1, 2004; 35(7): 1756 - 1762. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Shi, J. Macinko, B. Starfield, J. Xu, and R. Politzer Primary Care, Income Inequality, and Stroke Mortality in the United States: A Longitudinal Analysis, 1985-1995 Stroke, August 1, 2003; 34(8): 1958 - 1964. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||